Hepatocellular carcinoma (HCC) is the fastest growing cause of cancer-related death in the United States. It is critically important to identify those at high risk for HCC and institute effective surveillance strategies for early diagnosis. Livr cirrhosis is the main risk factor for HCC. Ultrasound and ?-fetoprotein (AFP) every 6 months remains the surveillance modality most frequently used on cirrhosis patients but the sensitivity and specificity of the tests are low. Our extensive prior biomarker studies in HCC cases, cirrhosis controls and HCC pre-diagnosis samples identified novel markers for HCC risk prediction and surveillance. In this proposal, we will further evaluate the performance of these novel markers together with FDA-approved markers. The validation study will be performed in a large contrast MRI surveillance cohort that provides a unique opportunity to study biomarkers on clinical material from patients rigorously classified as having a very early disease in a surveillance setting. Longitudinal collection of blood samples will be particularly valuable in assessing how early biomarkers become positive during the period when lesions are observed to a confirmed diagnosis of HCC. Cirrhotic patients under contrast MRI surveillance (n=3500 patients) will be enrolled at Baylor-St. Luke's Medical Center and Houston Methodist Hospital. Data management, specimen repository and molecular assays will be performed at MD Anderson Cancer Center. It is anticipated that at least 150 HCCs will be detected during the study with most of them detected at an early stage. We will identify the panel of biomarkers that best distinguishes between early HCC and cirrhosis and that could therefore have utility in HCC surveillance. We will evaluate the capacity of the novel marker panel to detect preclinical disease and determine longitudinal changes in these biomarkers predictive of HCC development. Our prior studies also identified specific fatty acids that discriminate between patients with cirrhosis who developed HCC within 5 years and those who didn't progress to HCC. These fatty acids may therefore have utility in HCC risk assessment. We will determine using the same MRI surveillance cohort the fatty acid panel that best predicts HCC development. Finally, this study will evaluate the complementarity performance of ultrasound and the novel biomarker panel and determine whether contrast-enhanced ultrasound results in any significant improvement in the ability of ultrasound to detect small tumor foci. This proposal achieves in one study two major goals: early detection and characteristics of tumors when biomarker becomes positive. This study has a number of high-impact translational applications: spare patients from unnecessary imaging tests; identify at-risk patients and trigger the decision to perform MRI instead of ultrasound in patients under surveillance; detect lesions at an early stage allowing for curative treatment. Such clinical applications would significantly reduce the cost of HCC surveillance and improve survival of HCC patients.